Policy-making by governments affects the behaviour of large numbers of people, sometimes millions. So why is such a key task often left to economists and lawyers, who may have little in-depth understanding of how people really behave? And why are the behavioural psychologists and neuroscientists who have valuable expertise usually consulted last, if at all?

Some politicians recognize this problem, and have tried to address it in the past few years. Leading the way is the British government under Prime Minister David Cameron, who established a Behavioural Insights Team (BIT) within the influential Cabinet Office shortly after he was elected. This summer, the White House Office of Science and Technology Policy began hiring people for the US equivalent.And on 30 September, some 300 people — including leading executives in governments, businesses, non-govern­mental organizations and academia — will gather in Brussels to discuss how behavioural insights can inform policy-making.

The conference has been organized by the European Commission’s Directorate General for Health and Consumers, which relies on the insights of behavioural scientists it has recruited in recent years. For instance, the choice of the next library of pictorial and text warnings on cigarette packets will be based on tests of their effectiveness, such as tracking people’s eye movements when they view them. The European Commission introduced a set of such warnings in 2005, but they were tested only with highly biased declarative methods, such as surveys and self-reporting.

The lesson here is that if a governing body such as the European Commission — infamous for its bureaucracy and resistance to change — can see the benefits of behavioural insights and alter its way of informing policy, then every government should follow its lead.

At the heart of this approach are the randomized controlled trials that are already common in medical research. Similar trials of public policy are crucial because they use a control group — a fraction of the population to which the new policy is not applied. This might sound strange, but monitoring such a non-intervention group is the only way to know whether a change in behaviour is down to the policy being trialled.

The British government is streets ahead on this. When the BIT was founded in 2010, the French Prime Minister’s Centre for Strategic Analysis was already running a programme to chart the benefits of using behavioural and brain sciences to inform public policy, which I had led for a year. But over the two years that followed, while we published reports with the hope of convincing our administration to give our field studies the green light, the British team was running trials and getting results.

Public organizations can no longer afford not to test policies before they are applied.

One successful example of a cost-effective use of behavioural insights in policy is a UK study on tax collection. In a 2011 randomized controlled trial of more than 100,000 people, some people received payment-request letters that had been tweaked to say that most UK citizens pay their taxes on time. Compared with control letters, the trialled policy produced a 15% increase in repayment rate. The British government estimates that a national roll-out of the policy would provide around £30 million (US$48 million) of extra revenue each year. Not bad for a smart use of social psychology.

Randomized controlled trials can help governments to choose the best strategy, to spot errors and thus to save a lot of money. The private sector has long understood that. Take OPower, a company headquartered in Arlington, Virginia, that promotes the sustainable consumption of energy. It reports that its clients have saved more than 2.8 billion kilowatt hours of energy as a result of its clever blend of trials, insights from social psychology and behavioural economics — also known as nudges, after the title of a 2008 best-selling book. The company found that the best way to get people to save energy was not to make them feel guilty about the environment or to promise cheaper bills, but to provide households with an easy way to compare their energy consumption with that of their neighbours. Social comparison and peer pressure work better than any lecture on how one should behave.

Many of us who work in the field of behavioural change have received educational training and research grants paid for by public funds. Yet the private sector seems to be much more interested in our skills than are public institutions, despite the high return that we could provide on investments in behavioural-sciences research. During these difficult economic and social times, public organizations can no longer afford — financially and socially — not to test policies before they are applied.

I am sure that most readers, regardless of where they live, could share examples of failed policies that have seen public money wasted on education, health, law, transport and taxes — all because no one cared to assess how citizens might behave when new measures are implemented. No one would accept that a new drug would be developed only by economists and lawyers and launched without the proper trials. We should not tolerate this in policy-making either.